Manuel Tonneau

HateDay: Insights from a Global Hate Speech Dataset Representative of a Day on Twitter

2024 · arXiv (Cornell University)

To address the global challenge of online hate speech, prior research has developed detection models to flag such content on social media. However, due to systematic biases in evaluation datasets, the real-world effectiveness of these …

When Claims Evolve: Evaluating and Enhancing the Robustness of Embedding Models Against Misinformation Edits

2025

Online misinformation remains a critical challenge, and fact-checkers increasingly rely on claim matching systems that use sentence embedding models to retrieve relevant fact-checks.However, as users interact with claims online, they often introduce edits, and it …